re: st: determining between subject basic unit in repeated measures

[David Airey <david.airey@vanderbilt.edu>]

.

You are missing a factor for subject. Try the following commands:

sort group block
egen person = seq(), from(1) to(36) block(2)
order group person block
list
table person block, by(group) concise c(n anticipation mean  
anticipation)
anova anticipation group / person|group block group*block, repeated 
(block)

which produce:

                           Number of obs =      72     R-squared      
=  0.7516
                           Root MSE      = 7.48311     Adj R-squared  
=  0.4655

                  Source |  Partial SS    df       MS            
F     Prob > F
            ------------- 
+----------------------------------------------------
                   Model |  5590.36839    38  147.114958        
2.63     0.0029
                         |
                   group |  602.019231     2  301.009616        
2.45     0.1020
            person|group |   4058.0229    33  122.970391
            ------------- 
+----------------------------------------------------
                   block |  57.2271728     1  57.2271728        
1.02     0.3194
             group*block |  873.099089     2  436.549544        
7.80     0.0017
                         |
                Residual |  1847.89994    33   55.996968
            ------------- 
+----------------------------------------------------
                   Total |  7438.26833    71  104.764343


Between-subjects error term:  person|group
                     Levels:  36        (33 df)
     Lowest b.s.e. variable:  person
     Covariance pooled over:  group     (for repeated variable)

Repeated variable: block
                                          Huynh-Feldt epsilon         
=  1.0625
                                          *Huynh-Feldt epsilon reset  
to 1.0000
                                          Greenhouse-Geisser epsilon  
=  1.0000
                                          Box's conservative epsilon  
=  1.0000

                                            ------------ Prob > F  
------------
                  Source |     df      F    Regular    H-F      G- 
G      Box
            ------------- 
+----------------------------------------------------
                   block |      1     1.02   0.3194   0.3194    
0.3194   0.3194
             group*block |      2     7.80   0.0017   0.0017    
0.0017   0.0017
                Residual |     33
            ------------- 
+----------------------------------------------------


see the FAQ at

http://www.stata.com/support/faqs/stat/anova2.html

specifically:

http://www.stata.com/support/faqs/stat/anova2.html#wsanova2

-Dave
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/